Partnership for Carbon Accounting Financials (PCAF & Climate Risk Modelling)

The Partnership for Carbon Accounting Financials (PCAF) is the global standard that banks and investors use to measure their financed emissions — the greenhouse gases tied to the money they lend and invest. And that measurement is the first, foundational step in building climate risk into credit risk models.

Why? Because you can't manage, price, or stress-test a risk you haven't measured yet.

In other words, PCAF tells you how much carbon sits inside your loan book. Climate credit risk modelling tells you what that carbon could cost you. One feeds the other.

If you work in credit, risk, lending, portfolio management, sustainability & ESG strategy or treasury, this is a core skill rather than a "nice to have". Regulators are no longer asking politely.

Let's explore why.

What is the Partnership for Carbon Accounting Financials (PCAF)?

PCAF is an industry-led initiative. It was created in 2015 by 14 Dutch financial institutions that wanted a consistent way to measure the emissions they were financing.

The idea caught on, and it expanded to North America in 2018, went global in 2019, and became a non-profit in 2023. Today, more than 650 financial institutions across five continents use it.

At its heart sits the Global GHG Accounting and Reporting Standard for the Financial Industry — usually just called the PCAF Standard. It gives banks, asset managers, and insurers a shared recipe for calculating the carbon footprint of their loans and investments.

These financed emissions are classified as Scope 3, Category 15 under the Greenhouse Gas (GHG) Protocol. Here's the part that surprises people.

For most financial institutions, financed emissions make up the vast majority of their total carbon footprint. Often more than 90%, and for some banks over 99%.

One study found that the emissions a bank finances are, on average, around 700 times larger than the emissions from its own offices and operations.

Let that sink in. A bank could switch every light bulb in every branch and barely move the needle.

The real carbon story is in the lending.

PCAF isn't standing still, either. In December 2025, it rolled out a major update, expanding its methods to ten asset classes and adding new guidance on forward-looking metrics.

Financed Emissions: A Quick, Friendly Definition

Think of financed emissions like this. If you lend a company money, and that company pumps out CO₂, you "own" a slice of those emissions proportional to your share of their financing. It's a bit like a fitness tracker for your loan book, except instead of counting steps, it counts carbon.

The basic formula is refreshingly simple:

Financed emissions = the borrower's emissions × your attribution factor

The attribution factor is your slice — for example, your loan as a share of the company's total value.

So why should a credit risk modeller care about any of this?

Because carbon-heavy borrowers tend to carry transition risk. The risk that new climate rules, carbon prices, or shifting customer demand eat into their profits. And a borrower whose cash flows shrink is a borrower more likely to miss a payment.

That's the bridge. Carbon is no longer just an ethics conversation. It's a credit quality signal.

How Climate Risk Enters a Credit Risk Model

Let's talk numbers. This well-known formula is what most banks use to estimate the Expected Loss (EL) on a loan:

Expected Loss = EAD × PD × LGD

Here's what those three letters mean:

• EAD (Exposure at Default): how much is owed if the borrower defaults.
• PD (Probability of Default): the chance they actually default.
• LGD (Loss Given Default): how much you'd lose after recovering what you can.

Climate change pushes on all three. It does so through two main channels:

The Basel Committee's Principles for the effective management and supervision of climate-related financial risks expect banks to fold both risks into PD, LGD, and EAD. Yet research from the IMF suggests most credit risk models still don't fully price climate risk in.

Translation? There's a real gap between what regulators expect and what most banks can currently do. And gaps like that are exactly where skilled people get hired.

Where PCAF Fits: Measure First, Model Second

This is the key insight:

You cannot build climate factors into a credit model if you don't know which borrowers are carbon-intensive in the first place. PCAF gives you that map. It turns a vague worry ("we probably have some climate exposure") into hard numbers ("here's our financed emissions by sector, by client, and by asset class").

PCAF is also refreshingly honest about data. It uses a data quality score from 1 to 5, where 1 is the best (based on real, reported emissions) and 5 is the weakest (based on rough economic estimates).
Most banks score the bulk of their portfolios in the 3-to-5 band, simply because so few borrowers report verified emissions yet.

Why does this matter for credit risk?

Because the quality of your carbon data sets a ceiling on the quality of your climate-adjusted PD and LGD. Better data in, better risk estimates out. Garbage in… well, you know the rest.

Example 1: The ECB Climate Stress Test — A Wake-Up Call

In 2022, the European Central Bank (ECB) ran its first climate risk stress test on 41 of the banks it directly supervises. Under a tough scenario combining a disorderly transition with physical risks like floods and droughts, those banks faced combined credit and market losses of around €70 billion.

And the ECB was clear: that number understates the real risk. The test covered only about a third of the banks' exposures and assumed no broader economic downturn.

Two findings stand out for anyone in credit:

1. Almost two-thirds of the banks' income from corporate customers came from greenhouse-gas-intensive industries.
2. Much of that financed-emission exposure was concentrated in a small number of large counterparties — meaning the risk wasn't spread out; it was bunched up.

Perhaps most telling was that at the time, around 60% of banks had no climate stress-testing framework, and only about 20% factored climate risk into their lending decisions at all.
The pressure hasn't eased since.

In the 2025 EU-wide stress test, transition-risk shocks added roughly 74 basis points of extra credit losses on corporate loans under the adverse scenario, with the biggest hits landing on banks most exposed to energy-intensive sectors.

The message from supervisors is hard to miss. Measure your financed emissions, understand your exposure, and build it into your models — or expect some very pointed questions.

Example 2: ABN AMRO and the Carbon Hiding in Its Mortgages

Now for a more hopeful story. ABN AMRO, one of PCAF's founding Dutch banks, used the PCAF method to measure the carbon impact of its lending. When the numbers came back, its mortgage portfolio turned out to be one of the largest sources of its financed emissions.

So what did it do? Rather than panic, it acted. The bank began promoting lending products that encourage energy-efficient homes — better for the planet, and a smart way to manage long-term risk.

This is PCAF working exactly as intended: measurement leading to action. And it connects neatly to physical risk, too. Picture a flood hitting a region full of mortgaged homes. Property values drop, collateral weakens, and LGD climbs. Floods, unfortunately, don't check your credit policy before they arrive.

A bank that has already mapped where its riskiest, least energy-efficient properties sit is far better placed to respond.

Why This is a Career Skill: Not Just a Compliance Chore

Let's connect the dots. The world's largest banks have funnelled staggering sums into high-carbon activity. According to the annual Banking on Climate Chaos report, the world's biggest banks have poured more than $7.9 trillion into fossil fuels since the Paris Agreement was adopted in 2015 — that's well over $2 billion every single day.

Regulators have noticed. The data and modelling demands are only growing. And yet the number of professionals who can confidently connect financed emissions to PD, LGD, and EAD is still small.
That's a skills gap. And skill gaps are opportunities for the people who close them first.

If you can walk into a room and explain how PCAF data flows into a climate-adjusted credit model, you're not just ticking a regulatory box. You're making yourself genuinely valuable and the kind of professional who gets the interesting projects, the promotions, and a seat at the table.

Reading about climate credit risk is a great start. Doing it confidently, and in line with what the ECB and Basel actually expect, is what sets you apart.

Redcliffe Training's Integrating Climate in Credit Risk Modelling course takes you from the core concepts straight through to the practical steps. In just one focused day, you'll learn how physical and transition risks transmit into PD, LGD, and EAD, how to run climate scenario analysis using NGFS and IEA scenarios, and how to apply the very tools supervisors look for — including PCAF, PACTA, and science-based targets. Better still, it's taught by an expert who has contributed directly to the ECB's climate stress test.

Don't wait for the next regulatory deadline to force your hand — get ahead of it. Become the person your team turns to when climate risk lands on the agenda.

FAQ

What is the purpose of Part A in the Partnership for Carbon Accounting Financials Standard?

Part A of the PCAF Standard covers Financed Emissions. Its purpose is to give financial institutions a transparent, consistent set of methodologies for measuring and disclosing the greenhouse gas emissions linked to their loans and investments — known as Scope 3, Category 15 emissions.

Following the December 2025 update, Part A spans ten asset classes, from business loans and mortgages to listed equity, project finance, and motor vehicle loans. In short, it's the rulebook that helps banks and investors calculate the carbon footprint of what they finance, in a way that's comparable across the industry.