Thanks for being a paid subscriber to Climate Risk Review! You can read this, and other articles from the archive, on your browser here. The best way to support the newsletter is by letting others know about it. Please consider sharing the below article with your network here:

Share

Interested in a group subscription for your whole team? Get in touch via louie.woodall@climateriskreview.com

It’s the rallying cry of climate disclosure enthusiasts the world over: “You can’t manage what you can’t measure”. The mantra encapsulates the belief that climate risks cannot be addressed and mitigated until they’re properly identified and quantified. The trouble is getting agreement on what to measure — and how to measure it.

It’s a particularly knotty debate within the banking sector. Climate-conscious stakeholders want to know lenders’ exposures to climate-related risks, and what these look like system-wide. This is, in fact, one of the goals of the Task Force on Climate-related Financial Disclosures (TCFD) — the climate reporting framework sponsored by the Financial Stability Board. Identifying these exposures is tricky, though. Getting widespread agreement across institutions on what does and does not constitute a climate risk-sensitive asset is even harder.

The TCFD itself took a shot at a baseline definition. It recommended that banks disclose their ‘carbon-related assets’, which it identified as exposures tied to the energy and utilities sectors under the Global Industry Classification Standard, excluding water utilities and independent power and renewable electricity producer industries. At the same time it muddied the waters by admitting that the term ‘carbon-related’ is a hazy one at best. Not exactly a ringing endorsement.

Perhaps in part because of this, disclosure of ‘carbon-related assets’ by banks has yet to take off. Climate Risk Review reviewed the disclosures of 117 banks that had signed up as TCFD supporters as of November 2020, and found that only 24 referenced ‘carbon-related assets’ specifically. Even fewer actually disclosed these as recommended by the TCFD — both as an amount and as a percentage relative to total assets.

‘Carbon-related assets’ of TCFD supporting banks

In addition, plenty of reporting institutions have veered away from the concept of ‘carbon-related assets’ entirely in favour of their own methods for totting up climate risk-sensitive exposures. In a recently-published TCFD report playbook — a sort of blueprint for disclosures put together by the UN Environment Programme Finance Initiative (UNEP FI), the Institute of International Finance and EY — no fewer than 12 climate risk metrics relevant to banks were identified. Ideas on how to classify exposures by their climate risk-sensitivity have clearly become more diffuse, not less.

This bodes ill for those stakeholders seeking comparability across institutions on their climate exposures. “The TCFD objective was to harmonise the corporate disclosure framework,” says Jakob Thomä, managing director of the 2° Investing Initiative Initiative (2DII), which oversees the Paris Agreement Capital Transition Assessment (PACTA) methodology.  “And what they’ve achieved [instead] is that everyone you know uses the same words, but the coherence of the actual disclosures is dramatically different from the common accounting frameworks and reporting definitions that underpin them”.

On the straight and narrow

One issue with the TCFD definition of ‘carbon-related assets’ is its narrowness. It therefore lowballs the quantity of exposures that could suffer financial loss because of climate change. As the above table shows, ‘carbon-related assets’ make up a vanishingly small portion of reporting banks’ exposures.

Furthermore, it’s a sector-based definition detached from traditional financial risk measures. These typically grade exposures using probability-of-default (PD) and loss-given default (LGD) variables, which are generated using historical data.

Small wonder there’s a push to broaden it out. The TCFD playbook argues that the financial sector should agree on “a broader view of sensitive sectors, beyond what the TCFD initially defined as carbon-intensive industries” in order to capture all exposures vulnerable to physical and transition risks.

Certain banks have already taken this leap. Citi, in its 2020 TCFD report, published a ‘heat map’ of exposures to sectors and subsectors that face high transition and physical risk using an in-house methodology. The transition risk of a sector, for example, was judged by the emissions intensity of their products, carbon intensity of their operations, and the likelihood or existence of climate-related policies affecting them. Valerie Smith, chief sustainability officer at Citi, says the bank views this heat map as “a living document”, which will adjust in response to new data and changes to the practices, technologies and policies of each sector.

UK lender Barclays also disclosed its own proprietary measure of exposures subject to “elevated risk” from climate change, placing this alongside its report on ‘carbon-related assets’ as defined by the TCFD. Doing so illustrated the gap between its in-house measure of climate exposures and the TCFD’s baseline. Barclays’ ‘carbon-related assets’ amounted to £27 billion at end-2019. “Elevated risk” exposures, in contrast, totalled £36.7 billion.

Efforts to expand the concept of climate risk-sensitive exposures are being pursued by financial watchdogs, too. The European Banking Authority (EBA) trialled two approaches of its own last year, the results of which were published in its December Risk Assessment Report. The first used climate policy relevant sectors (CPRS), a classification system cooked up by European academics back in 2017, which is based on projections of those industries most likely to be disrupted by the transition to a low-carbon economy. The second was an emissions-based methodology that utilised carbon intensity data as a proxy for transition risk.

These pilot exercises may inform the EBA’s push to create harmonised ESG risk disclosures for banks, draft regulatory standards for which should be out later this year. “The disclosure proposed by the TCFD is a good one, but it’s true that the definition of carbon-related exposure is quite narrow,” says Pilar Gutierrez, senior policy expert at the EBA. “What we are trying to do is define a template through which banks can disclose information on carbon-related exposures by sector, [either by] using emissions data provided by Eurostat or based on the Climate Benchmark Regulation that splits sectors into high-climate and low-climate impact sectors”.

The regulator is also monitoring the debate over whether the European Union should introduce an “unsustainable” activities list to complement its sustainable finance taxonomy. Such a list could shape future disclosure requirements, as the EBA could mandate supervisees publicly report their exposures to these unsustainable activities once they’re defined. The European Commission is due to report on the proposed taxonomy by the end of this year. “On the EBA side, we would very much welcome it if the Commission eventually decides to extend the taxonomy in order to provide criteria for the identification of climate-harmful or neutral activities,” says Guitierrez.

The watchdog may be waiting a long time, though, as plenty of political obstacles stand in the way. After all, activities tagged by the “unsustainable” badge could become climate pariahs and have their funding choked off. Lawmakers may be leery of ‘outcasting’ certain industries this way, fearing political retribution or socioeconomic fallout. Banks too may be reluctant to disclose their own breakdown of climate risk-sensitive assets out of concern that doing so could shine too bright a spotlight on certain clients — who’d likely not appreciate the attention.

The EBA’s efforts to date are also still predominantly sector-based, rather than risk-based. The CPRS methodology, for instance, identifies climate risk-sensitive exposures by mapping them to existing NACE economic activity identifiers, rather than by estimating climate-adjusted PDs or LGDs.

“When you have a granular dataset with detailed sector information, the CPRS approach has the advantage of being quite straightforward to be applied. However, working at detailed sector level to assess climate risk has some limitations, as for instance there might be different activities within the same NACE sector [that are more or less climate sensitive],” explains Raffaele Passaro, bank sector expert at the risk analysis and stress testing unit of the EBA.

In the interests of simplicity and comparability such an approach makes sense. Attempting to identify climate risk-sensitive assets using traditional financial risk measures is, after all, complicated by the fact that historical data on climate-related losses that could be used to adjust PDs and LGDs does not currently exist. Instead, forward-looking estimates have to be produced. Dozens of climate risk tools are working to produce these, but their very number — and very different approaches — means it is unlikely that any single one will be adopted as an industry-standard baseline any time soon.

Comparability vs ownership

But what sometimes gets lost in the debate is that cross-sector comparability will most likely come at the expense of bank-level utility. To put it another way, the further apart a harmonised categorisation of climate risk-sensitive exposures gets from a bank’s own understanding of its vulnerabilities, the less useful it will be for measuring and managing them.

“Before we decide whether we want that comparability we need to understand what the cost/benefits are, because there’s also the downside of comparability, which is a loss of ownership. The more comparable you make things, the more you’re going to get into a situation where people are going to say: ‘I don’t agree with this standard’,” says 2DII’s Thomä.

It’s a tension that’s shaped risk-based banking regulation before and since the global financial crisis. Under the incoming Basel III capital framework — named for the Basel Committee on Banking Supervision — lenders are subject to a harmonised “standardised approach” which sets capital requirements the same for all firms. However, they’re also allowed to use their own internal models, which generate institution-specific risk measures that can influence these to a limited degree.

This ‘best of both worlds’ approach is supposed to foster a degree of comparability while conceding that banks themselves know more about their risks than arms’ length supervisors. In a climate risk context this does, however, necessitate agreement on a baseline in the first place. The risk-insensitivity of the current TCFD ‘carbon-related asset’ definition suggests it’s not fit for purpose.

The question is whether any market-led or third-sector-led initiative could attain the requisite following to fulfill this role. Thomä’s PACTA methodology has accrued a number of influential supporters, chief among them the so-called ‘Katowice banks’ — five top European lenders — who have used it to assess the alignment of their credit portfolios with a Paris-aligned temperature pathway.

“Our hope [with PACTA] is to create a common floor through a bottom-up exercise. So we’re making it open source and IP rights free … to circumvent some of the competition around this. That creates what we’re seeing now with the Katowice banks, where we feel like we’re getting to a place when you look at those banks disclosures you start to be able to compare apples to apples,” says Thomä.

One obvious issue with PACTA acting as a baseline, though, is that it’s an alignment methodology, not a risk methodology. The difference is subtle, but important. Put simply, it cannot be used to produce climate-adjusted PDs or LGDs.

Ultimately, regulatory standard-setters, such as the Basel Committee and the FSB, will likely have to step in. If the objective of climate disclosure is really to further understanding of the financial system’s exposures to climate-related risks, then a truly risk-based measure is needed. It may be the case, then, that the TCFD’s ‘carbon-related asset’ metric is quietly retired, in recognition that as a means to count up and compare banks’ climate risks it simply can’t do the job.

Please send questions, feedback and more to louie.woodall@climateriskreview.com

You can catch climate risk management updates daily on LinkedIn

The views and opinions expressed in this article are those of the author alone

All images under free media license through Canva