Financed emissions (Scope 3 Category 15), explained
For banks and investors, Category 15 usually dwarfs every other emission source. Here is how financed emissions work, how PCAF attribution is calculated, and why data quality decides the number.
For most companies, the largest line in a greenhouse gas inventory is energy or a supply chain. For a bank, an asset manager or an insurer, it is almost always the money itself. The emissions tied to lending, investing and underwriting sit in a single Scope 3 category, and they tend to be very large. CDP found that the emissions associated with financial institutions' portfolios are on average over 700 times greater than the emissions from those institutions' own operations. If you run a financial firm's inventory and you have not measured this category, you have not really measured your footprint at all.
This article explains what Category 15 covers, how the Partnership for Carbon Accounting Financials (PCAF) turns a loan or a shareholding into an emissions number, and why the quality of the underlying data, not the formula, is where most of the work lives.
What Scope 3 Category 15 actually covers
The GHG Protocol Corporate Value Chain (Scope 3) Standard splits a company's value chain emissions into 15 categories. Category 15, Investments, captures emissions associated with a reporting company's investments that are not already counted in Scope 1 or Scope 2. It is designed primarily for financial institutions and for any company that makes investments as a service to others. In plain terms: if your money is working somewhere, the emissions produced there belong, in proportion, to you.
The Scope 3 Standard tells you what to account for. It does not give a step-by-step method for a mortgage book or a corporate bond portfolio. That gap is what PCAF fills. The PCAF Global GHG Accounting and Reporting Standard for the Financial Industry provides asset-class-specific methods that the GHG Protocol has reviewed as being in conformance with its requirements.
The attribution factor: the core idea
Financed emissions rest on one principle. You claim a share of an investee's or borrower's emissions equal to your share of their financing. That share is the attribution factor. Multiply it by the counterparty's own emissions and you have your financed emissions for that exposure.
Financed emissions = attribution factor x counterparty emissions, where the attribution factor is your outstanding amount divided by the counterparty's total value (for listed equity and corporate debt, PCAF uses enterprise value including cash, or EVIC).
The choice of denominator matters. For listed equity and bonds, PCAF's methodology uses enterprise value including cash so the same company value is used across all its investors, which keeps the attributed emissions from being double counted or lost. For business loans to private companies, total equity plus debt is used instead. The point is consistency: everyone financing the same company should, in aggregate, account for that company's emissions once.
The six asset classes
PCAF's first standard set out methods for six asset classes, and later versions extended the coverage. The core six are listed below.
- Listed equity and corporate bonds
- Business loans and unlisted equity
- Project finance
- Commercial real estate
- Mortgages
- Motor vehicle loans
Later PCAF guidance adds methods for sovereign debt, and for capital market activities such as facilitated emissions from underwriting. Each asset class has its own way of defining the outstanding amount and the counterparty emissions, but the attribution logic is the same throughout.
Data quality is the real work
The formula is simple. Getting a trustworthy counterparty emissions figure is not. Most borrowers and many investees do not report verified emissions, so financial institutions estimate. PCAF built a data quality scoring scale so that every reported number carries a signal of how it was produced.
| PCAF score | Basis for the emissions estimate | Certainty |
|---|---|---|
| 1 | Verified emissions reported by the counterparty | Highest |
| 2 | Unverified emissions reported by the counterparty | High |
| 3 | Physical activity data (energy use, production) with emission factors | Medium |
| 4 | Economic activity data (revenue, asset value) with sector factors | Lower |
| 5 | Estimates from asset turnover or region and sector averages | Lowest |
The scale runs from 1 (highest certainty, verified data) to 5 (lowest certainty, broad averages), as set out in PCAF's standard and in joint PCAF and CDP guidance on data quality. Moving an exposure from a score of 4 or 5 toward 2 or 3 usually means getting real energy and activity data for the counterparty rather than inferring emissions from revenue. That is precisely where clean utility data changes the answer: a borrower's actual metered consumption, structured and validated, is the difference between a spend-based guess and a physical-activity estimate.
Why disclosure rules are forcing the issue
Category 15 is no longer optional to think about. The ISSB's climate standard, IFRS S2, requires companies to disclose Scope 1, 2 and 3 emissions measured in accordance with the GHG Protocol, and it calls out that financial institutions should disclose additional information about their financed emissions. The GHG Protocol has confirmed that IFRS S2 requires disclosure of Scope 3 emissions. For a bank or investor, Scope 3 means Category 15, and Category 15 means PCAF.
Start with the exposures that dominate your book and the counterparties you can get real energy data from. A smaller portfolio measured at PCAF score 2 or 3 tells you more than a full book estimated at score 5.
Where MartinAI fits
Financed emissions live or die on the counterparty data underneath the attribution factor. When that data comes from utility bills, meter reads and Green Button feeds, it usually arrives messy: mixed formats, estimated reads, and gaps. MartinAI turns that raw material into clean, validated, analysis-ready activity data, so the physical-activity path (a PCAF score of 3 rather than a 5) is actually available to you. You keep the attribution logic and the reporting; we make sure the inputs are defensible.
Frequently asked questions
What is Scope 3 Category 15?
Category 15 (Investments) is the GHG Protocol Scope 3 category that captures emissions associated with a company's investments, lending and underwriting that are not already counted in Scope 1 or Scope 2. It applies mainly to financial institutions and is where financed emissions are reported.
How are financed emissions calculated?
You calculate an attribution factor, your share of a counterparty's financing, and multiply it by that counterparty's own emissions. PCAF defines the specific denominators and methods for each asset class, such as enterprise value including cash for listed equity and bonds.
What is the PCAF data quality score?
PCAF uses a scale from 1 to 5 to flag how an emissions figure was produced. A score of 1 means verified reported emissions, a score of 3 means physical activity data with emission factors, and a score of 5 means broad sector or region averages. Lower scores mean higher certainty.
Do disclosure rules require financed emissions?
IFRS S2 requires Scope 3 disclosure measured against the GHG Protocol, and for financial institutions that includes financed emissions. Many jurisdictions are adopting or referencing IFRS S2, which is pushing Category 15 from voluntary to expected.
- 1CDP: finance sector's funded emissions over 700 times greater than its own
- 2PCAF: Global GHG Accounting and Reporting Standard for the Financial Industry
- 3GHG Protocol: Corporate Value Chain (Scope 3) Standard
- 4PCAF and CDP: the importance of data quality
- 5GHG Protocol: ISSB IFRS S2 requires Scope 3 disclosure
- 6PCAF: about the partnership
GHG inventory boundaries: operational vs financial control
Before you count a single tonne, you have to decide what counts as yours. This is how the GHG Protocol organizational boundary works, and how the control approaches change your number.
Carbon accounting software still needs clean utility data
Carbon accounting platforms apply emission factors well. They do not fix bad inputs. Here is why activity data quality, not the software, decides whether your emissions number holds up.