‘Still a significant gap’: How government and financial institutions can account for the GDP impacts of climate change

In the UK, there’s now a growing recognition that many economic forecasts, made by the Treasury and financial regulators, overlook the future impacts of climate change. Within these forecasts, potential ‘scenarios’ are put forward – estimates of how climate change will impact on different demographics, industries, and GDP in the future.

The ‘Emperor’s New Climate Scenarios’ report, published in 2023, sounded alarm bells for forecasters in financial services, finding current estimates “significantly underestimate” the economic impacts of climate change. This warning, alongside other key 2023 publications including ‘No Time To Lose’ (published with USS) and ‘Climate Scorpion‘ (in collaboration with the Institute and Faculty of Actuaries), has sparked the development of new, alternative climate scenarios.

What’s the latest?

Most government departments and financial institutions base their analysis on scenarios from the Network for Greening the Financial System (NGFS). Until recently, standard NGFS scenarios predicted that 3.5°C of warming by 2100 would reduce global output by 7-14% – far less damage than climate scientists would expect to see at this temperature level.

As of November 2024, the scenarios now use an updated ‘damage function’ and estimate a 30% hit to global GDP by the end of the century – if we follow our current trajectory to reach just under 3 degrees of warming.

While this does a much better job of capturing economic damages, it still fails to capture the full range of climate risks and impacts. The estimates ignore several important factors, including: direct effects of heatwaves, rising sea levels, damage from tropical cyclones, and potential climate tipping points. They also fail to account for non-economic impacts of climate change, particularly to ecosystems and biodiversity, as well as human health and wellbeing.

What’s wrong with current economic models?

Ultimately, this comes back to the inherent flaws in using traditional ‘equilibrium’ models for climate economics. Fundamental uncertainty in climate change, such as fat-tailed risks, irreversible tipping points, and intricate feedback loops that are not accounted for by conventional probability distributions, is what they fundamentally fail to account for.

The underlying assumptions of these models—perfect markets, rational actors, seamless economic changes, and exogenous technology change—all falter in the face of climate change. Additionally problematic is their over-reliance on discount rates, which can drastically shift policy recommendations while essentially disregarding potentially disastrous long-term effects. They frequently have arbitrary damage functions that are unable to deal with severe situations, cascading failures, or important non-economic effects like societal instability and biodiversity loss.

Perhaps most troublingly, these frameworks typically ignore crucial questions of equity and justice, aggregating impacts in ways that obscure who causes emissions versus who bears their consequences.

This all stems from a fundamental misapplication: trying to force climate change into an optimisation framework when, in reality, it is a complex problem of risk management that involves deep uncertainty, ethical choices, and political economy that requires different analytical tools.​​​​​​​​​​​​​​​​

How can we build better models in future?  

More realistic models paint a dramatically different picture. Cambridge Econometrics, for instance, estimates a 65% GDP loss under a 4°C warming scenario, while other models suggest a 50-60% downside to existing financial assets in a business-as-usual scenario. These higher estimates better align with climate science and account for some of the crucial factors that current government models often ignore – including tipping points, compound risks, non-linear effects, social and political instability, mass migration impacts, and supply chain disruptions.

Data on these factors is now widely available. Landmark reports such as Global Tipping Points 2023 and ‘No Time to Lose: New Scenario Narratives for Climate Action‘ are just two recent examples that address the information deficit, providing crucial data on the tipping points, compounding effects and worst-case scenarios that need incorporating into climate modelling.

What needs to happen next?

Promisingly, there are signs of further change. The Government’s Actuary Department (GAD) recently advised the Treasury to adopt more realistic scenarios that include both 2°C and 4°C warming pathways, while accounting for tipping points and compound risks. They’re also pushing for more sophisticated analysis that quantifies rather than merely describes potential impacts, considers longer timeframes out to 2100, and updates their analysis more frequently – every 3-5 years. The Bank of England is also beginning to acknowledge these limitations and working to develop more comprehensive stress-testing approaches.

Yet, despite this increasing recognition of the problem within Whitehall, there remains a significant gap between the economic models being used for government planning and what climate science suggests are realistic impacts. This disconnect presents real risks for government planning and financial stability if decisions continue to be based on overly optimistic impact assessments.

While there appears to be movement toward more realistic assessment methods, this transition is still in its early stages and the use of benign – and potentially dangerous – impact estimates remains widespread across government departments and financial institutions.

Why does this matter?

This situation is particularly concerning given recent extreme weather events, like those in Spain, which demonstrate how climate impacts are already exceeding many traditional model predictions. To compound the problem, extreme weather events are occurring ahead of schedule, with greater severity, and at a frequency that surpasses most scientific projections from just a few years ago. These sorts of extreme weather events aren’t at all captured in the current approaches to the economics of climate change.

The response now is critical. Without rapid changes, the continued use of optimistic economic models – those that ignore tipping points and extreme weather events, and underestimate damages – could set a precedent for systemically underestimating climate risks in government policy and financial planning.

About the Author

Dr Jesse F Abrams is a climate systems expert at the University of Exeter and an Impact Fellow at Green Futures Solutions. To find out more about this work, contact him here.

Let’s Work Together

Green Futures Solutions is the University of Exeter’s initiative for business, connecting organisations with over 1,500 leading climate scientists to drive positive, sustainable change. Contact our expert team today.