The European Central Bank released its long-awaited inaugural banks’ climate stress test (CST) early July. This was an important first step for banks’ integration of climate risks in their internal models and mainly served as an exploratory exercise. Even so, some market participants are frustrated by the lack of concrete results. The latter are published in aggregate format only, and do not include capital depletion scenarios, as we are used to seeing in the regular banks stress-testing exercise. The results of the CST will only feed qualitative elements into the Supervisory Review and Evaluation Process (SREP). This means that results could only have an indirect effect on Pillar 2 capital requirements through SREP scores.
Still, we can be optimistic, as this exercise will be beneficial for all banks’ investors in the next iterations, as it will push banks to improve their climate risk modelling and disclosure.
SCENARIO:
The CST evaluates aspects of both physical and transitional risks. The assessment of physical risk focuses on two extreme weather events: a major flood and a severe drought over a one-year time horizon.
For transitional risks, the test includes several scenarios and time frames. First, it assesses banks’ short-term vulnerabilities on a three-year baseline and a disorderly transition scenario, triggered by a sharp increase in the price of carbon emissions. Second, it looks how their longer-term strategies deal with three different transition scenarios over a 30-year horizon:
- The orderly scenario: climate policies are introduced early on and global warming is limited to 1.5°C.
- The disorderly scenario: new climate policies are introduced only from 2030 onward and global warming is limited to 2°C.
- The hot house world scenario assumes that no new climate policies are implemented and that the current global efforts are insufficient to halt significant global warming. Global emissions rise until 2080, leading to global warming of around 3°C.
The exercise considers the impact of transitional risk from a credit, market, operational and reputational risk standpoints as well as from a qualitative perspective.
RESULTS:
A total of 104 significant banks participated in the test consisting of three modules, in which banks provided information on their: (i) own climate stress-testing capabilities, (ii) reliance on carbon-emitting sectors, and (iii) performance under different scenarios over several time horizons.
The bottom-up stress test within the third module was limited to only 41 banks. The stress test shows that credit and market losses in the short-term disorderly transition and the two physical risk scenarios amount to around EUR 70 billion on aggregate for the 41 banks in question.
Module 1:
Even though they’ve made some progress since 2020, the results show that banks do not yet sufficiently incorporate climate risk into their stress-testing frameworks and internal models. Around 60% of banks do not yet have a climate risk stress-testing framework and most of those banks envisage a medium- to long-term timeframe only for incorporating physical and/or transition climate risk into their framework. Similarly, most banks do not include climate risk in their credit risk models, and only 20% consider climate risk as a variable when granting loans.
Also, banks often rely on proxies to estimate their exposure to emission-intensive sectors, but various proxy techniques or data sources greatly influence the reported data, leading to deviations in reported scope emissions for the same counterparty across banks as we can see in the chart below :
