PCAF Financed Emissions: How to Calculate Category 15 Scope 3 for Banks and Investors

PCAF Financed Emissions: The Standard That Changed How Banks Measure Scope 3

Banks, asset managers, and insurers face a unique challenge in GHG accounting: their Scope 1 and 2 emissions — the energy used to run their offices and data centers — are trivial relative to the emissions of the companies and projects they finance. A regional bank might have 2,000 tCO2e from its own operations and $12 billion in commercial loans to energy-intensive industries. The emissions that matter are Scope 3 Category 15 — financed emissions — and calculating them correctly requires the PCAF Standard.

The Partnership for Carbon Accounting Financials published its Global GHG Accounting and Reporting Standard for the Financial Industry in 2020 (updated 2022, Part C covering insurance added separately). It defines asset-class-specific attribution factors, acceptable data sources ranked by a 1-5 quality score, and the disclosure format that TCFD, CSRD ESRS E1, and the Net-Zero Banking Alliance all reference. For a credit analyst or sustainability officer at a financial institution, understanding PCAF is now a baseline competency — not optional ESG knowledge.

ClaudeFinanceLab's ESG & Sustainable Finance templates include PCAF Financed Emissions Calculator and CSRD ESRS E1 Disclosure Assistant, both designed for finance teams doing their first PCAF inventory or improving data quality scores across an existing one.

The Attribution Factor: How Banks Claim Their Share of Borrower Emissions

The core PCAF methodology is simple: for each borrower or investee, calculate what fraction of the company's total financing the bank/investor provides (the attribution factor), then multiply that by the company's total emissions. The resulting number is the bank's "financed emissions" from that exposure. Aggregate across the portfolio to get total Category 15 Scope 3.

Attribution factors differ by asset class because the economic relationship between financier and borrower differs:

• Corporate loans and bonds: Attribution factor = Outstanding Amount / (Total Equity + Total Debt). Rationale: the bank's loan is one component of the entire capital structure that funds the company's operations.
• Listed equity: Attribution factor = (Share of equity owned × EVIC) / EVIC = percentage equity ownership. Or equivalently: Market Value of Investment / EVIC, where EVIC = Enterprise Value Including Cash.
• Project finance: Attribution factor = Outstanding Loan / Total Project Cost. Projects (power plants, infrastructure) have self-contained financing where the lender's share of project cost is clear.
• Commercial real estate: Attribution factor = Outstanding Loan / Property Value. Coverage is on the asset, so the property value is the denominator.
• Mortgages: Attribution factor = Outstanding Loan / Property Value. Individual properties, but PCAF provides building energy intensity data (kWh/m² by property type and country) to estimate emissions where utility data is unavailable.
• Motor vehicle loans: Attribution factor = Outstanding Loan / Vehicle Value. Emissions calculated from vehicle type and estimated annual km driven.
• SME loans: Attribution factor = Outstanding Loan / (Total Equity + Total Debt). Same as corporate but data availability is much lower — most SMEs don't report emissions.

• "PCAF attribution factor calculation for a corporate loan portfolio: We have 3 commercial loans. Loan 1: $85M outstanding to Apex Manufacturing (Scope 1+2 = 145,000 tCO2e, Total Equity $280M, Total Debt $420M). Loan 2: $32M outstanding to Redwood Logistics (Scope 1+2 = 38,000 tCO2e, Total Equity $95M, Total Debt $145M). Loan 3: $15M outstanding to Pinecrest Retail (Scope 1+2 = 12,000 tCO2e, Total Equity $42M, Total Debt $28M). For each loan: (1) calculate the PCAF attribution factor, (2) calculate financed emissions attributed to our bank, (3) calculate financed emissions per million dollars of outstanding loan. Total portfolio financed emissions?"
• "Listed equity PCAF calculation: Our equity portfolio holds positions in 4 public companies. Company A: we hold 2.4M shares, share price $48.50, total shares outstanding 180M, EVIC = $12.4B, Scope 1+2+3 = 2.8M tCO2e. Company B: we hold 850K shares, share price $124.00, total shares outstanding 420M, EVIC = $68B, Scope 1+2 = 890,000 tCO2e (Scope 3 not disclosed). Company C: we hold 5.1M shares, share price $22.30, total shares outstanding 95M, EVIC = $2.8B, Scope 1+2 = 340,000 tCO2e. Company D: we hold 1.2M shares, share price $215.00, total shares outstanding 310M, EVIC = $85B, Scope 1+2+3 = 18.6M tCO2e. Calculate financed emissions for each position and total portfolio. Note which companies have Scope 3 data and the difference if we include/exclude Scope 3."
• "PCAF project finance financed emissions: We are the lead arranger in a $480M project finance facility for a natural gas combined cycle power plant (1.2 GW capacity, 65% capacity factor). Our ticket: $95M. Total project cost: $680M (remainder is equity and other debt). Annual electricity generation: ~6.1 TWh. Emission factor for natural gas combined cycle: ~0.34 kgCO2e/kWh (EPA eGRID 2024). Calculate: (1) annual project emissions in tCO2e, (2) our PCAF attribution factor, (3) our financed emissions from this project, (4) financed emissions per MW of installed capacity for comparison with clean energy alternatives, (5) what would financed emissions be if this were a wind farm with the same capital structure?"

PCAF Data Quality Scores: Ranking Your Evidence

PCAF's most important operational concept is the data quality score — a 1-5 scale that tells you (and your disclosures) how reliable your emissions estimate is. Score 1 requires third-party verified reported data. Score 5 is a spend-based estimate with no company-specific data at all. You must report the weighted-average data quality score for each asset class — it tells readers whether your financed emissions figure is precise or a rough approximation.

In practice, most bank portfolios look like this today: large-cap listed companies in regulated sectors (Score 1-2), mid-market borrowers who report to CDP or in their annual report (Score 2-3), SMEs and privately held companies with no disclosed emissions (Score 4-5). The disclosure requirement is not to have perfect data — it is to know what you have and report the quality honestly.

• "PCAF data quality scoring for our commercial loan portfolio: We have 240 corporate borrowers. Breakdown of emissions data availability: 18 borrowers have third-party verified GHG data (Scope 1+2) reported to CDP — these are our largest exposures averaging $95M outstanding. 47 borrowers disclose Scope 1+2 in their sustainability reports (unverified, but company-reported with methodology notes). 65 borrowers are in sectors where sector emission intensities are available (e.g., steel, cement, chemicals) — we can estimate using PACTA/IEA sector intensities × their revenue. 110 borrowers are private SMEs with no emissions data — we use EEIO-based spend/revenue estimation. How should we assign PCAF data quality scores to each group? What is our weighted-average DQ score if these 4 groups represent 35%, 28%, 22%, and 15% of outstanding balance respectively?"
• "Improving PCAF data quality — engagement strategy: Our financed emissions report has a weighted-average data quality score of 3.8 — mostly Score 4 sector averages and Score 5 EEIO. Our NZBA (Net-Zero Banking Alliance) commitment requires a data quality improvement roadmap. Top 20 borrowers by outstanding balance represent 60% of our financed emissions. Of those: 8 already disclose (Score 1-2), 7 are CDP-eligible but not currently reporting, 5 are private companies that will never file CDP. Design an engagement strategy: (1) what information should we request from the 7 CDP-eligible non-reporters? (2) what data can we collect for the 5 private companies? (3) what improvement in weighted-average DQ score is realistic in 12 months? (4) draft the outreach language for the 7 CDP-eligible borrowers."

Financed Emissions by Sector and Temperature Score

PCAF financed emissions by themselves don't tell you whether your portfolio is aligned with climate targets. For that, you need sector-level analysis — which industries account for most of your financed emissions — and temperature alignment metrics that link portfolio emissions to a warming trajectory. The two main frameworks for temperature scoring are PACTA (Paris Agreement Capital Transition Assessment) for forward-looking technology alignment, and TCFD/PCAF financed emissions for backward-looking inventory.

• "Sector concentration analysis for financed emissions: Total portfolio financed Scope 1+2 emissions: 4.2M tCO2e from $8.4B in corporate loans. Sector breakdown of financed emissions: Power generation (utilities) 1.8M tCO2e ($1.2B outstanding), Oil & Gas (upstream) 890K tCO2e ($680M outstanding), Steel and metals 620K tCO2e ($420M outstanding), Cement 340K tCO2e ($210M outstanding), Commercial real estate 280K tCO2e ($2.1B outstanding), Other 270K tCO2e ($3.79B outstanding). Calculate: (1) financed emissions intensity per million dollars of exposure by sector, (2) which sector is most emissions-intensive per dollar lent, (3) what % of financed emissions come from fossil fuel-related sectors (power + O&G), (4) how does our power sector intensity compare to IEA NZE 2050 pathway (which requires power sector emissions to reach near-zero by 2050)?"
• "Portfolio decarbonization target calculation: We committed to net-zero financed emissions by 2050 under the NZBA. Current financed emissions: 4.2M tCO2e (2024 baseline). Interim target: 50% reduction by 2030 = 2.1M tCO2e by 2030. Our portfolio composition will change: we project $400M in new renewables lending annually, $200M/year reduction in fossil fuel exposure (natural run-off, not forced divestment), and our existing borrowers need to decarbonize. If existing borrowers achieve the IEA NZE Announced Pledges Scenario pathway (reduces sector emissions 35% by 2030 on average), and we execute the new lending plan: (1) what is our projected 2030 financed emissions, (2) are we on track to meet the 2.1M tCO2e target, (3) what additional actions would close any gap?"

CSRD ESRS E1 Disclosure Requirements for Financial Institutions

Under CSRD ESRS E1 (Climate), financial institutions subject to the directive (large EU banks, insurers, and investment firms; non-EU institutions with EU subsidiaries above thresholds) must disclose Scope 3 Category 15 financed emissions. The required disclosures include: total financed emissions by asset class, weighted-average PCAF data quality score by asset class, percentage of portfolio with SBTi-aligned targets, and the institution's own net-zero target and interim milestones. CSRD ESRS E1 explicitly references PCAF as the preferred methodology for financial institutions.

• "CSRD ESRS E1 financed emissions disclosure: Draft the financed emissions section of our ESRS E1 disclosure. Our data: Total financed Scope 1+2 emissions: 4.2M tCO2e (PCAF methodology, 2024 reporting year). Asset class breakdown: Corporate loans $8.4B outstanding / 3.1M tCO2e, Listed equity $2.1B AUM / 680K tCO2e, Commercial real estate $1.8B / 280K tCO2e, Project finance $420M / 140K tCO2e. Weighted-average PCAF data quality scores: Corporate loans 3.4, Listed equity 2.1, CRE 4.2, Project finance 1.8. % of financed emissions from borrowers with SBTi-validated targets: 22%. Our net-zero target: 2050, interim 50% reduction by 2030 vs 2024 baseline. Draft the ESRS E1 disclosure paragraphs covering: methodology, asset class table, data quality commentary, limitations, and target alignment."
• "PCAF mortgage portfolio financed emissions: We have a residential mortgage portfolio of $3.2B outstanding across 18,400 loans. We do not have individual energy consumption data for most properties. PCAF mortgage methodology allows property-type-based emission intensity estimates. UK average residential emission intensity by property type (BEIS 2024): Detached house 4.2 tCO2e/year, Semi-detached 3.1 tCO2e/year, Terraced 2.8 tCO2e/year, Flat 1.9 tCO2e/year. Our portfolio mix by property type (by number of properties): Detached 28%, Semi-detached 35%, Terraced 22%, Flat 15%. Average outstanding loan-to-value: 68% (average LTV, which means attribution factor = LTV / 100 per PCAF). Average property value: $285,000. Calculate: (1) weighted average annual emissions per property, (2) attribution factor, (3) total financed emissions from the mortgage portfolio, (4) financed emissions intensity per million dollars outstanding, (5) what PCAF data quality score does this approach receive?"

Insurance-Associated Emissions (PCAF Part C)

The 2022 update to the PCAF Standard added Part C, covering insurance underwriting — the idea being that an insurer who underwrites a coal plant is "enabling" those emissions through risk transfer in the same way a bank that lends to it enables emissions through capital provision. The attribution factor for insurance is: Annual Premium / (Annual Revenue of insured). This is the most contested part of PCAF — some large insurers dispute the economic logic — but EIOPA and the Net-Zero Insurance Alliance (NZIA) have both referenced it.

• "Insurance underwriting financed emissions — PCAF Part C: We are a commercial insurer. Relevant underwriting portfolios: Property insurance for a cement producer: annual GWP $4.2M, insured's annual revenue $280M, insured's Scope 1+2 = 1.8M tCO2e. Marine cargo insurance for an oil trading company: annual GWP $890K, insured's annual revenue $2.1B, insured's Scope 1+2+3 = 14.2M tCO2e. Directors & Officers insurance for a renewable energy developer: annual GWP $340K, insured's annual revenue $95M, insured's Scope 1+2 = 28K tCO2e. Calculate insurance-associated emissions for each policy using the PCAF Part C attribution factor. What % of our insurance-associated emissions comes from the oil trading policy? What is the data quality challenge here?"

Where to Start

The practical starting point for most banks is the corporate loan portfolio — it's usually the largest asset class by balance and the one where emissions data is most accessible for larger borrowers. Run the PCAF attribution factor calculation for your top 20-30 borrowers by outstanding balance — those typically represent 50-60% of financed emissions. Score each one on the 1-5 data quality scale. Use the ClaudeFinanceLab ESG & Sustainable Finance templates to draft your PCAF methodology documentation and ESRS E1 disclosure narrative.

See also: GHG Emissions Calculator with Claude AI and CSRD and ESG Reporting with Claude.

Frequently Asked Questions

Is PCAF required or voluntary for banks?

PCAF itself is voluntary — it is an industry standard, not a regulatory requirement. However, it is the methodology referenced by the Net-Zero Banking Alliance (NZBA) for the 140+ signatory banks, by the Task Force on Climate-related Financial Disclosures (TCFD), and explicitly by CSRD ESRS E1 for EU financial institutions. For banks subject to CSRD, Scope 3 Category 15 disclosure is mandatory — and PCAF is the only widely-accepted methodology for calculating it. So while "PCAF" isn't legally required, the underlying financed emissions disclosure it enables increasingly is.

What is EVIC and why does it matter for PCAF equity attribution?

EVIC is Enterprise Value Including Cash — market cap plus total debt plus minority interest, minus cash. It is the denominator in the PCAF attribution factor for listed equity and bonds: Attribution = Investment Value / EVIC. EVIC fluctuates with market prices, which means a bank's financed emissions from listed equity change even without any portfolio changes — purely due to share price movement. PCAF recommends using year-end EVIC values for comparability. The market sensitivity is a known limitation; some institutions also report financed emissions normalized by EVIC to reduce this noise.

How should we handle Scope 3 emissions in PCAF calculations — include them or not?

PCAF recommends including Scope 3 emissions of the borrower/investee where data is available and material, particularly for: oil and gas (Scope 3 Category 11 — use of sold products — dominates the footprint), automotive (Scope 3 Category 11 — tailpipe emissions from vehicles sold), agriculture/food (Scope 3 Category 1 — purchased goods). For sectors where Scope 3 is less material (services, financial sector borrowers), Scope 1+2 may suffice. The data quality score applies: Scope 3 from CDP (Score 2) is better than no Scope 3 at all, but Scope 3 estimated via EEIO (Score 5) adds more noise than signal for most portfolios.

What is the difference between absolute financed emissions and financed emissions intensity?

Absolute financed emissions is the total tCO2e attributed to a financial institution's portfolio — it grows as the portfolio grows, even if each borrower decarbonizes. Financed emissions intensity normalizes by portfolio size (tCO2e per million dollars of assets or loans outstanding) — it can improve even if absolute emissions grow due to portfolio expansion. Both metrics are useful: NZBA interim targets typically use absolute emissions for power and fossil fuel sectors (where absolute reduction is the goal) and intensity metrics for sectors where the economy needs to continue to grow (e.g., infrastructure). CSRD ESRS E1 requires disclosure of both.