Why UK banks can't model their way out of a climate data problem and the solution is already available to them.

When the UK's biggest banks gathered recently to address inconsistencies in their climate risk disclosures, the urgency was understandable. But the conversation needs a shift in focus. Major financial institutions don't have a modelling problem; they face a data infrastructure challenge. And even the most sophisticated model cannot fully compensate for fragmented baseline data.

I've sat with enough Chief Risk Officers to know how complex this journey is. Even after internal teams are mobilised and data gaps acknowledged, risk models relying on siloed inputs can still produce estimates that vary significantly from reality. In fact, in tests based on traditional sector level estimation models, client data was shown to be up to 200% out of line when compared to available company level emissions data. By contrast, by using more granular data sets, the Equifax Financed Emissions Calculator was able to narrow that variance to 15% or less¹. This data gap is not a reflection of internal capability; it is entirely a reflection of data access.

As climate risk is increasingly factored into Expected Credit Loss (ECL) models at the loan level, relying on broad industry averages or siloed internal data can create challenges during audit reviews. The Prudential Regulation Authority’s (PRA) SS5/25 supervisory standards highlight the growing importance of this.

The challenge is structural. A bank's sustainability team has access to whatever its borrowers choose to disclose, supplemented by sector-level estimates. A Credit Reference Agency sees the whole picture: who lends to whom, exact loan amounts, credit health of the borrower, who the company is on a micro sector level and payment behaviour of every business in the portfolio. That's why Equifax sits at the centre of this, as the orchestration layer that connects the credit data banks already rely on, with the climate data they urgently need. Without that foundation, frameworks remain reliant on estimation.

Banks have been treating climate disclosure as an Environment, Social and Governance (ESG) reporting challenge, something for the sustainability function to solve rather than a credit risk challenge. The borrowers most exposed to transition risk - like businesses reliant on old fuel systems, or commercial property owners with large retrofit and transition liabilities, or landlords with a buy-to-let portfolio non-compliant with the upcoming Minimum Energy Efficiency Standards (MEES) - are the same borrowers who are at significant risk of default with the ongoing energy price spikes, or their assets become un-rentable. As an example, a bank holding just 100 commercial properties that miss that Energy Performance Certificate (EPC) deadline could be looking at a potential £20m–£40m retrofit capital shock, when considering the average size of a SME commercial office at 2000sq ft with £400k average investment per building². That lands on the balance sheet, not the sustainability report.

Developed in partnership with our sustainability data partner CienDos, we have a full suite of data and insight products that directly address the climate challenges financial institutions are facing. CienDos tracks carbon exposure at individual borrower level rather than applying broad sector averages, achieving an average Partnership for Carbon Accounting Financials (PCAF) score of 3.18. Our Property Transition Intelligence product models the real financial consequences of embarking (or not embarking) on a retrofit investment. Furthermore, the Sustainability Assessment Manager tracks how changing regulations and energy costs will affect individual companies globally, feeding directly into ECL calculations.

Combined with Equifax core commercial bureau data, banks can move from guesswork to an accurate, forward-looking view of transition risk across their portfolios. The same data that identifies exposed borrowers also identifies which ones urgently need capital to upgrade, a qualified green lending pipeline that may not otherwise be visible. Banks treating SS5/25 as a compliance cost to minimise are missing that second part. The transition creates a lending opportunity. Businesses across the UK need capital to decarbonise, upgrade buildings, and adapt their operations. Lenders with accurate data will see that demand, understand the unique risks, and price it correctly before their competitors do.

The data required to achieve this already exists. By leveraging these existing datasets rather than recreating them internally, banks can potentially achieve a commercial advantage over their competitors, navigate regulatory expectations, and better position themselves to capitalise on market opportunities.

Sources:
¹ Equifax & CienDos Environmental Data Intelligence report - 2025
² Calculation Averages, Frank Knight research, Meeting the Commercial Property Retrofit Challenge - Part 2: The Business Case for Action - Link
Maths breakdown : 2,000 sq ft: The size of an average, mid-sized commercial office. £200 per sq ft: The current industry cost (via Knight Frank) to deep-retrofit an old building to meet the mandatory 2027 EPC 'C' legal standard. £400,000: The total upgrade cost for just one building. £40,000,000: The sudden capital shock if a bank holds just 100 of these non-compliant properties in its loan book.