A letter was sent by Senators Toomey, Shelby, Crapo, Scott, Rounds, Tillis, Kennedy, Hagerty, Lummis, Moran, Cramer, and Daines to Chairman Jerome Powell of the US Federal Reserve, dated March 18 which quotes our work. We seek to clarify some substantive issues.

We write concerning the letter described above, which was tabled in the Senate Committee on Banking, Housing and Urban Affairs. One of the letter’s central assertions is that climate-related financial risk information is subject to “substantial methodological challenges associated with assessing climate-related risks that undermine the usefulness” of assessing banks’ climate risk.   

The letter cites two academic papers to support this claim, including our paper, Business risk and the emergence of climate analytics, published by Nature Climate Change in February. The letter cites our paper in this sentence:

“As researchers have noted, current climate models provide little financially meaningful information.”

This citation entirely misrepresents our analysis and recommendations.

While our paper highlights potential misuses of climate models for financial analysis it also notes numerous ways in which climate models do provide value in their current form, provided data from climate models are correctly interpreted and applied. Below, we list examples of where our paper explicitly makes these observations.

We also published a plain language summary of take-away points specifically for busy decision-makers such as Senator Toomey and his colleagues, who may not be versed in climate science but for whom the paper might prove useful.

The summary makes clear that our critique is nuanced (emphasis added):

Our paper could be misinterpreted as advising that no-one should use climate models, outside of the model developers and a few academics with deep knowledge of them. That would be an unfortunate and incorrect reading of our message. In fact, we note that the current research-driven approach to climate modelling is prone to exaggerating the difficulties for business to extract actionable information that is scientifically defensible.”

We also note that the letter then switches from discussing the physical models that are the subject of our paper, and of the Hawkins et al (2015) paper that is subsequently cited, to argue about “stranded assets”. Stranded assets is a financial hypothesis concerning market responses to climate mitigation, that bears no relation to the impacts addressed by our paper and Hawkins et al. As the letter does not cite any further research to back up its claim concerning “stranded assets” risk, we are concerned the authors are confusing the two main forms of “climate-related financial risk” — “physical risk” and “transition risk” – a distinction that has been recognised by many financial regulators and authorities around the world, including the Financial Stability Board’s Taskforce on Climate-related Financial Disclosure (TCFD) and the Network on Greening the Financial System (NGFS).

In addition we wish to point readers to an example in the public domain, the Commonwealth Bank of Australia’s 2019 Annual Report, where climate change science was effectively integrated with detailed metrics from the Australian Bureau of Agricultural and Resource Statistics to provide a forward looking risk scenario of the agriculture and regional banking portfolio. In contrast to the concerns raised in the letter, banks regularly include forward looking estimates in their annual reports, including in their consolidated accounts, that are uncertain in nature.

Our paper suggests some ways that government, industry and research sectors could better develop and utilise climate modelling capabilities. Our expertise spans different facets of climate change and finance, but we all understand that the imperative to do this is urgent.


Dr Tanya Fiedler
Prof Andy Pitman
Kate Mackenzie
Dr Nick Wood
Dr Sarah Perkins-Kirkpatrick
Prof Christian Jakob

Selected statements in our paper that highlight where climate models can be effectively used:

“The climate science community provide centennial-scale simulations for many models, which can provide a rich resource for business. Where global or continental-scale projections of average temperature or the direction of change in rainfall for 2050 or 2100 are required, for different emission scenarios, climate models are essential tools.

“Analysis of the observational record, combined with observed impacts on a portfolio and expert judgement from the weather and climate sciences of future changes, can provide useful guidance over the next decade or two.

“Although GCMs and downscaling can guide decision-making, a case-by-case approach informed by a deep understanding is essential of where and when climate projections provide useful information.“Nonetheless, in terms of the provision of global temperature change, and excluding the impact of global scale tipping points, GCMs forced by a range of RCPs or SSPs provide useful input for financial and economic risk modelling techniques at global and continental scales.