This is the second article in a three-part series exploring the European Banking Authority (EBA) Pillar 3 ESG regulations and what they mean for banks’ climate and ESG data strategies. In the first article, we explained the origins of these regulations and provided an overview of how the disclosure requirements are organized into ten templates provided by the regulator. These disclosure templates broadly fall into four buckets:
In this article, we examine the first two areas: transition and physical risks related to the disclosure of an institution’s ‘banking book’. This is the assets on a bank’s balance sheet intended to be held until maturity, typically in the form of loans to retail and corporate clients, as well as derivative instruments primarily used for risk management. We will also identify the data banks require to complete templates 1-5 and explore potential data sources and approaches they can leverage in this pivotal evolution of the climate risk disclosure framework.
Under EBA Pillar 3, transition risk refers to banks having exposures to firms or assets that are at risk of losing some or all of their value – or even becoming liabilities – as a result of changes associated with the global transition to net zero. The most obviously risky exposures are directly to companies whose business model depends on high-carbon activities, but there are more subtle cases, such as companies that consume or sell large amounts of plastic, or agribusiness enterprises that use huge quantities of fertiliser produced with natural gas-based ammonia. Different sectors can be affected in many ways by more stringent climate policies, accelerating technological change and, increasingly, legal and reputational risks.
To gauge the amount of transition risk a given bank is exposed to, regulators need to understand in detail how a bank’s balance sheet looks when broken down into sectoral exposures and the absolute greenhouse gas emissions (Template 1), or associated energy or emissions intensities (Templates 2 and 3), attributable to each one. As a catch-all proxy for risk, Template 3 asks banks to measure the alignment between the emissions intensities of their exposures and those consistent with the International Energy Agency (IEA) Net Zero Emissions by 2020 (NZE) scenario, while Template 4 looks at exposures to a list of the world’s largest carbon emitters.
Within prudential disclosure frameworks, climate risks have until now been considered as effectively just another component of general risk management. The new Pillar 3 regulations now require banks to identify and quantify the effect of climate change on the risk profile of their exposures. As one of the most advanced such regulatory regimes globally, Pillar 3 will certainly push banks to gather, analyse and explain more climate-specific data than most of them have done to date. With the EBA’s focus being on financial stability, compliance may result in tighter capital controls – but it may also help banks subject to the regulation develop a deeper and clearer understanding of their exposure to climate risks than their non-EU competitors, while also focusing minds and resources on building up exposures to sectors that stand to gain from the transition, thereby helping to funnel capital in the direction of the transition.
In narrative terms, banks must explain qualitatively how they are appropriately managing risks associated with regulatory changes driven by climate policy and related targets. In some cases, policy expectations can be derived from explicit provisions in domestic or EU law, or from the results of specific legal cases. For example, the recent ruling by a district court in the Hague requiring Shell – a client of many of the banks regulated by Pillar 3 – to cut its emissions by 45% by 2030, relative to 2019. They are also required to explain their strategies for managing the risks of rapid, potentially disruptive technological shifts that may be policy-driven but also result from breakthroughs in the availability and affordability of clean technologies relative to high-carbon incumbents – such as growth in dispatchable renewable power replacing gas, and electric vehicles replacing the internal combustion engine and the associated demand for refined oil products.
In Template 1 (‘Credit quality of exposures by sector, emissions and residual maturity’), data inputs are grouped by the main economic activities in which counterparties are engaged (according to the NACE classification system already widely used in the EU and further afield). The template lists 51 categories designated by the European Commission as “sectors that highly contribute to climate change”; in other words, activities and technologies that are currently greenhouse gas intensive in one way or another.
For each of the listed NACE categories, banks must supply:
In Template 2 (‘Loans collateralised by immovable property – Energy efficiency of the collateral’), banks report on loans they have made that are collateralised by commercial and residential property (i.e. primarily mortgages), and collateral obtained through repossession. The key requirements are to provide average energy efficiency metrics, both in terms of energy consumption per square metre; and Energy Performance Certificate (EPC) labelling. Finally, banks report on collateral that does not have an EPC label, and the share of exposures for which energy efficiency levels are estimated rather than actual.
Template 3 (‘Alignment Metrics’) takes a closer look at bank exposures’ overall alignment with external emissions targets key carbon-intensive sectors, including power, fossil fuel combustion, automotive, aviation, maritime transport, cement, steel and chemicals (all of which are covered by Asset Impact). Alignment metrics differ by sector. Banks must choose from several possible metrics used by the IEA, then measure how far away they are from the IEA’s NZE scenario, and the corresponding short-term targets that would bring them in line with the NZE trajectory.
In Template 4 (‘Exposures to top 20 carbon-intensive firms), counterparty-level exposures are required for the twenty largest carbon-emitting companies globally. The EBA does not prescribe a single source for this list, allowing disclosing entities to base their list of top twenty emitters on “publicly available reputable and accurate information”.
Physical risk corresponds to banks’ exposures to firms or assets at risk of losing some or all of their value, or becoming liabilities, as a result of the physical impacts of climate change. This includes, but is not limited to, extreme heat and precipitation, extreme weather events, wildfire, drought, and sea level rise. There are obvious chronic (longer-term, ongoing) risks – for instance, companies with low-lying coastal properties – and acute (short-term) risks, such as companies with physical assets in known wildfire zones or flood plains. In addition, there are more subtle risks, such as increasing operating costs for companies that will require more cooling amid higher temperatures, and increased disruption to infrastructure from flooding events temporarily preventing or limiting the effective running of a business.
Understanding a banking book’s physical climate risk profile ideally requires splitting its exposures into specific risks, geographical zones, and sectors and weighting the potential value-at-risk (or equivalent metric) accordingly. In practice, this information can be very difficult to obtain at such a detailed level, and the definitions and scope of risks are debated. Interdependencies between different types of physical risks, and how these cascade down corporate supply chains, are in general well beyond the capacity of banks to report at present. To a significant degree, the Pillar 3 requirements on physical risk reflect these realities.
Partly in recognition of the difficulty inherent in the categorisation of physical risks, Template 5 (‘Exposures subject to physical risk’) requires only that banks provide financial information on the assets they hold which in geographical areas subject to acute and chronic physical risks, and to both. Exposure information must be provided separately for nine real economy sectors, and additionally for loans to (and repossessions of) residential and commercial properties. Similarly to Template 1, information on asset impairments, non-performing loans, and maturities is also required.
The data requirements for meeting the demands of Templates 1-5 are daunting. To a degree, financial databases that classify companies by their NACE codes, and corporate reporting of greenhouse gas emissions, can help to build a partial picture of a financial institutions’ climate risk exposure. But not all companies across all sectors report their emissions, especially for private companies not compelled to disclose by legislation or stock exchange listing rules. Still fewer companies report production figures reliably enough to calculate the physical emissions intensities and efficiency performance required by Templates 2 and 3. For disclosures under Template 5 – physical risk – to have any value, banks will need to use location-based data that assesses the physical risk factors affecting counterparties’ assets around the world.
The corporate emissions data industry exists in large part to meet these needs but tends to rely on a combination of backward-looking corporate disclosures complemented by statistical estimates to fill data gaps. The physical risk data industry, meanwhile, typically requires banks to know their counterparties’ locations in advance.
At Asset Impact, we bring all these requirements together in one place in an integrated solution for banks’ Template 1-5 data challenges. Our database covers over 330,000 physical assets linked to over 66,000 listed and unlisted companies at every level of the corporate ownership tree from subsidiaries to ultimate parents. It includes standard financial identifiers to help match the data to banks’ portfolios. Designed to be comparable across companies and over extended time horizons, our models are fully transparent on the methodologies used to associate emissions and emissions intensities both with individual assets and then to roll these up to the corporate level. This approach also enables us to build a forward-looking picture of the climate performance of every company in our data. It also means that we can offer geolocated asset-level data for most of the sectors we cover, allowing banks to secure a list of the assets owned by individual counterparties and to assess the physical risks affecting the corresponding locations, feeding directly into their data requirements for Pillar 3, as well as supporting them in internal target-setting and tracking.
Click here to read the final article in this three-part series, we turn to templates 6-10, the Green Asset Ratio and how the Pillar 3 regulations relate to the broader EU Taxonomy.
For more information on Asset Impact’s products and how our comprehensive and granular data solutions can support your own institution’s EBA Pillar 3 disclosure efforts, contact us via the button below.