Climate Stress Testing and the Cautious Approach of European Supervisory Authorities

This summer has seen heatwaves and a long period of drought hit large parts of Europe. Among the many risk and policy implications, this development puts a spotlight on the financial risks caused by climate change. This blogpost examines climate risk as a risk to financial stability or as a systemic risk. One nascent risk management tool that regulators and supervisors around the world are grappling with is climate stress testing.

While the rise in the use of climate stress tests is a global trend, our recent paper, forthcoming in Utrecht Law Review, focuses on the three European Supervisory Authorities (ESAs) in the European Union, ie, the European Banking Authority (EBA), the European Securities and Markets Authority (ESMA) and the European Insurance and Occupational Pensions Authority (EIOPA). The ESAs are increasingly active in the area of climate-related financial risks. Although sustainability matters might already be viewed as part of their original mandate, the revised ESA Regulations, introduced in 2019, have made these duties more explicit. Their mandates now expressly include assessing market exposure to systemic environmental risks and developing common methodologies for institutions to test their resilience to adverse environmental developments.

Our paper accordingly provides an overview of the ESAs’ work on climate stress testing against the broader backdrop of EU’s continuing process of agencification. It also examines the legal nature of the broad spectrum of soft law tools, such as guidelines and opinions, that ESAs have deployed in conjunction with climate stress tests. Our analysis reveals that the delegation of powers towards the ESAs appears to remain within the boundaries of the Meroni doctrine and its interpretations. The doctrine, gradually relaxed over time, raises some questions on the legitimacy of generally applicable acts by the ESAs, like those stemming from the exercise of soft law powers. Yet, within the purview of our analysis, we concluded that this concern should not be overstated. While ESAs’ mandates on climate risks would in principle leave room for the use of climate stress tests for micro-prudential or macro-prudential purposes, prior exercises have largely been exploratory. 

A Spectrum of Climate Stress Tests

There are two broad categories of climate stress tests: ‘top-down’, designed by supervisory authorities or central banks, and ‘bottom-up’, conducted at firm level. Whilst a detailed overview may be found in our paper, our analysis demonstrates varying degrees of ‘institutionalisation’ of climate stress tests across the ESAs’ practices. The three authorities have deployed climate stress tests, or required market actors to conduct their own, in different circumstances and to a different extent. Most of these exercises remain primarily analytical in nature, quantifying and assessing the magnitude of climate-related risks within a given market segment and requiring relatively limited follow-up by financial institutions. Only more recently have the ESAs begun issuing guidance with a regulatory function, prescribing how financial institutions should conduct their stress tests (as in the case of EBA pursuant to the revised Capital Requirements Directive). Either way, ESAs’ initiatives still appear to stay comfortably within the boundaries set by the non-delegation doctrine first articulated in Meroni. If anything, the ESAs could go a little further than they are currently doing and perhaps be a little more ambitious in setting their expectations.

Looking Ahead: More Room to Manoeuvre

As the methodology for the design of climate stress tests still requires fine-tuning, and climate-related policies are being rolled back globally (in the EU and beyond), it is challenging to predict the future uptake of such exercises across ESAs’ practices. That said, recent initiatives suggest a timid yet emerging emphasis on such practices. For example, the ESAs have jointly developed guidelines for national competent authorities, crystallising a number of best practices for the integration of environmental, social and governance risks within supervisory stress tests. In accordance with Art 16(3) of the ESA Regulations, these have a ‘comply-or-explain’ application, with some national competent authorities having already announced their intention to comply.

As another example, under the category ‘top-down’ exercises, EBA has recently published the draft methodology for its 2027 EU-wide stress test, conducted on a biennial basis. The design of the test itself, still subject to consultation, appears to take into account compliance costs for banks, alleviating the reporting requirements across a wide range of data points. Even so, the outcomes are expected to provide a more comprehensive assessment of climate risks than previous exercises. For the first time, the exercise will feature a dedicated climate risk module, which will require banks to disclose qualitative information on their exposure to both physical and transition risks. In order to assess these, banks will need to rely on scenarios identifying the potential economic damages of river floods and of transition-related shocks linked to carbon prices, GHG emissions, as well as gas and oil prices. For the time being, the draft envisages a separate consideration of such risks, going one step further in integrating them within banking supervision. Yet, they are not expected to affect the core stress test results. In other words, unlike other outcomes related to other risk drivers, a high exposure to climate risks will not have any negative repercussions on the Supervisory Review and Evaluation Process (SREP) scores and thereby on banks’ capital adequacy requirements. In principle, however, EBA’s mandate could leave room for bolder measures, eg tying their findings to micro-prudential or macro-prudential requirements (as others have also proposed). Other ESAs could likewise make a more pervasive use of these exercises: for instance, the 2026 ESMA’s report on Stress Testing for Central Counterparties or its consultation on stress testing for Money Market Funds fail to incorporate climate-related matters. 

Conclusions

Against the backdrop of the ESAs’ broad mandate on climate-related risks, our article invites reflection on the broader function that these exercises could serve. Broader use of climate stress tests, as well as linking their outcomes to the microprudential or macroprudential frameworks, would not amount to an overstepping of the ESAs’ mandate, but rather to a natural consequence thereof.

Admittedly the methodologies for these tests are not ripe yet, and market actors may lack some data to accurately report on climate-related risks. The approval of the Omnibus I directive, excluding several non-financial companies from disclosure obligations on sustainability matters, may amplify this challenge. Nevertheless, climate-related risks remain as material as ever, as the ‘Fit-for-55’ exercise also concluded. The stress test, jointly conducted by the ESAs and the European Central Bank, identified significant losses for financial institutions associated with transition risks when coupled with macroeconomic shocks. This justifies a broader, and even more efficient, use of climate-related stress tests, so that they do not remain toothless and symbolic exercises. It is precisely in times of simplification and reduction of compliance costs that it appears imperative for the ESAs to make full use of the mandates conferred upon them, elevating climate stress tests to consolidated practices, giving rise to real supervisory consequences.

The authors’ paper is forthcoming in Utrecht Law Review and available here.

Federica Agostini is an Assistant Professor in the International and European Law Department and an Affiliated Researcher at Utrecht Centre for Regulation and Enforcement in Europe (RENFORCE), Utrecht University, the Netherlands.

Marloes van Rijsbergen is a Senior Policy Advisor at the Dutch Authority for the Financial Markets, the Netherlands.

Ebbe Rogge is an Associate Professor at the Leiden University, the Netherlands and a Senior Policy Advisor at the Dutch Authority for the Financial Markets. 

The opinions expressed herein are solely those of the authors and in no way represent those of the Dutch Authority for the Financial Markets.